Stress is the demand placed upon an organism subjected to real or perceived threat or challenge. In order to maintain homeostasis, the organism mounts an array of hormonal, autonomic and behavioral responses, some of which are common to most stressful circumstances, such as activation of the pituitary adrenal axis and sympathetic nervous system and behavioral arousal. The stress response in large part is regulated through Corticotropin-Releasing Factor (CRF), a 41-residue hypothalamic peptide that stimulates the secretion and biosynthesis of pituitary ACTH, leading to increased adrenal glucocorticoid production. This process is regulated through a negative feedback loop whereby glucocorticoids suppress CRF production.
Although originally isolated and characterized on the basis of its role in this hypothalamopituitary-adrenal (HPA) axis, CRF has been found to be distributed broadly within the central nervous system as well as in extraneural tissues, such as the adrenal glands and testes, where it may also act as a paracrine regulator or a neurotransmitter. Moreover, the involvement of CRF in affective disorders, such as depression and anorexia nervosa, and in modulating reproduction and immune responses suggests that changes in CRF expression may have important physiological consequences. For example, perturbations in the regulatory loops comprising the HPA axis often produce chronically elevated levels of circulating glucocorticoids; such patients display the physical hallmarks of Cushing's syndrome, including truncal obesity, muscle-wasting, and reduced fertility. Most cases of Cushing's syndrome are caused by ACTH-producing tumors of the pituitary or, less frequently, nonendocrine tissue. Adrenal gland tumors or ectopic adrenal tissue account for 10-30% of occurrences of the disorder, but, in contrast to the pituitary-dependent form, plasma ACTH levels are not elevated. Several patients with Cushing's syndrome have been reported with ectopic CRF-secreting tumors, leading to the proposal that CRF can chronically drive pituitary ACTH production and, in turn, glucocorticoid release. It has been suggested that excess production of CRF may cause pituitary hyperplasia, leading to microadenoma formation and excess ACTH production. That pituitary hyperplasia accompanies some CRF-secreting tumors is consistent with this proposal.
CRF is thus a very potent stimulator of the synthesis and secretion of various peptides in the human body. The rat and human species have the same CRF molecule (r/h CRF or hCRF), which is a 41-residue peptide having the structure which is set forth in U.S. Pat. No. 4,489,163. Ovine CRF (oCRF) was first characterized, and its 41-residue structure is set forth in U.S. Pat. No. 4,415,558.
Although CRF levels in human peripheral circulation are normally low, there are often elevated levels of CRF in the maternal circulation, which levels progressively increase throughout pregnancy. It has been found that the increasing concentrations of CRF in pathological cases of pregnancy, such as pregnancy-induced hypertension and pre-term labor, are substantially and significantly elevated above those found in normal pregnancies (Campbell et al., J. Clin. Endocr. & Metab., 64:1054-1059, 1987).
It is believed that this maternal plasma CRF most likely originates from the placenta wherein it plays a paracrine role. Placenta cells have been shown to respond to CRF and to produce CRF and its mRNA. Even though CRF concentrations measured in late gestational maternal plasma are similar to levels reported in rat hypothalamic portal blood, which levels are capable of stimulating ACTH release in vitro, it does not appear that there is normally overproduction of ACTH during pregnancy. However, maternal plasma ACTH concentrations do increase slightly with advancing gestation.
A number of workers have used molecular crosslinking with radioiodinated CRF to identify putative ovine CRF binding proteins and receptors in brain, pituitary and AtT-20 cells which range from 40-70 kD in molecular weight. After subtraction of the molecular weight of the cross-linked CRF (.about.5 kD), the main protein form that was found in the pituitary gland was reported to be 70 kD, Nishimura et al., J. Biol. Chem., 262, 12893 (1987). A lower molecular weight protein of about 50 kD was reported to be the major brain form; Grigoriadis, et al., Endocrinology, 125, 3068-3077 (1989). The heterogeneity in sizes of these proteins was thought to be possibly due to differential glycosylation because, after N-glycanase treatment, only one cross-linked species of about 40-46 kD was observed in both brain and pituitary.
There were also reports of proteins in human plasma which are capable of biologically inactivating CRF, see Linton, E. A., et al. Clin. Endo. 28, 315-324 (1988) and Behan, D. P., et al. J. Endo. 122, 23-31 (1989), the latter of which discloses a partial purification process which resulted in an isolate that has now been determined to have been no more than about 50% pure. The purification of the isolated protein was ultimately accomplished, and sequencing of the pure compound provided sufficient amino acid sequence information to clone the DNA encoding this protein, which is now referred to as human serum hCRF-binding protein (hCRF-BP) [SEQ ID NO:6], E. Potter, et al., Nature, 349, 423-426 (Jan. 31, 1991). It has been proposed that the role of this protein substance might be the prevention of inappropriate pituitary-adrenal stimulation during pregnancy, and recombinant rat and human serum CRF-BPs have now been expressed in COS cells. They have been found to bind to the 41-residue CRF with high affinity, so as to be capable of therapeutically modulating the effect of CRF.
Some additional preliminary work has been done trying to isolate CRF receptors. Grigoriadis and DeSouza, J. Biol. Chem., 263, 10927-10931 (1988) and Grigoriadis, et al., supra, speculated that the molecular weight of the brain CRF receptor to be about 58 kD, less the MW of ovine CRF; however, their characterizations have been limited to estimates of the molecular weight of this protein by SDS-PAGE analysis of covalent complexes formed by chemical crosslinking between the receptor and .sup.125 I-CRF which are present within crude extracts containing a myriad of other proteins. They have not published more definitive information with regard to the CRF receptors and thus have not enabled others to determine or utilize the receptor structures.
Synaptic membrane CRF binding sites in the mammalian brain are integral to central relays for several sensory modalities including the olfactory bulb which comprises prominent sites of CRF-BP gene expression. The presence of membrane-associated CRF receptors in the mammalian brain are demonstrated inter alia by the presence of binding proteins in important cell groups which mask the immunodetection of CRF peptides present for the regulation of corticotropin production and intercellular communication of the central nervous system. Present studies in this field are aimed at determining how the expression of membrane-associated brain CRF-BPs are regulated by stress and corticosteroid influences.
In order to study the structure and biological characteristics of brain-derived membrane-associated proteins which bind to CRF and also to study the role played by these binding proteins in the responses of various cell populations to CRF stimulation, or to use them effectively in therapy, as components in affinity columns, diagnosis or assay, homogeneous compositions of the binding proteins are needed. Such compositions are theoretically available via purification of solubilized proteins expressed by cultured cells; however, even in cell lines known to express detectable levels of CRF receptors, such is present as a very minor component of total cellular proteins. It is therefore desirable that the nature and the structure of such membrane-associated CRF binding proteins be ascertained so that these proteins can be provided in sufficient quantity to allow them to be utilized for screening of compounds for drug design, for therapy by modulation of transactivation of CRF receptors by means of competition for tissue binding sites, for affinity columns and for other appropriate purposes.